Classifying Electrooculogram to Detect Directional Eye Movements

Banerjee, Anwesha; Datta, Shreyasi; Pal, Madhumangal; Konar, Amit; Tibarewala, D. N.; Janarthanan, R.

doi:10.1016/j.protcy.2013.12.338

Cited by 40 publications

(15 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The RBF or Gaussian kernel is defined by (11) where σ denotes the width of the Gaussian. The initial layer is the input layer and the final layer is the output layer.…”

Section: ) Power Spectral Densitymentioning

confidence: 99%

“…EOG based Eye movement controlled human computer interfaces are the major interests of recent HCI research. Several instances of EOG-based control in Human Computer Interactions are found in the literature [10][11][12], including controlling motion of computer cursor [13] and controlling wheelchair system for rehabilitation [14]. There have been different strategies of analyzing [15,18] and implementing EOG in the field of robotics [16].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Electrooculogram based blink detection to limit the risk of eye dystonia

Banerjee

Pal

Tibarewala

et al. 2015

2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)

Self Cite

View full text Add to dashboard Cite

In this paper a system for detecting the possibility of eye dystonia, a neural disorder that causes a person to blink excessively, by eye movement analysis is proposed. The designed system counts the number of blinks for a particular time interval and thus detecting the risk of eye dystonia. Electrooculogram (EOG) signal is recorded to collect eye movement data using a laboratory developed acquisition system. Radial Basis Function(RBF) kernel Support Vector Machine (SVM) classifier and Feed forward neural network classifier is used to classify blinks from other types of eye movements using combinations of Wavelet coefficients, Autoregressive (AR) parameters and Hjorth parameters with Power Spectral Density (PSD) as signal features.A maximum average accuracy of 95.33% over all classes and participants is obtained using RBF-SVM classifier with a feature space of AR parameters of order 5 and PSD taken together.

show abstract

“…The RBF or Gaussian kernel is defined by (11) where σ denotes the width of the Gaussian. The initial layer is the input layer and the final layer is the output layer.…”

Section: ) Power Spectral Densitymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Electrooculogram based blink detection to limit the risk of eye dystonia

Banerjee

Pal

Tibarewala

et al. 2015

2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Dynamic time warping (DTW) was implemented to tackle the duration difference between the reference and testing template. Banarje et al [11] tried to extract three EOG features, which were wavelet coefficients, Power Spectral Density (PSD),and Auto Regression (AR) model and combination of these, and classified them by two methods, K-nearest neighbor (K-NN) and feed forward neural network method. This research concluded that neural network classifier reached better accuracy than K-NN.…”

Section: Introductionmentioning

confidence: 99%

Controlling 3-D Movement of Robot Manipulator using Electrooculography

Rusydi¹,

Okamoto²,

Ito³

et al. 2018

IJEEI

View full text Add to dashboard Cite

EOG is a biosignal generated by eye activities. This research develops the use of EOG signals in a new way. EOG was used to control a robot manipulator that reaches for a target in 3-D area. Five properties of EOG signal were used, which were: threshold, polarity of signal, integral of absolute EOG signals, peak and period of a wave. The use of these properties allowed the apparatus to detect the EOG signals, estimate gaze distance and distinguish between gaze motion, voluntary blinks and involuntary blinks. Targets were recorded by two cameras that provided front and top views. The images were viewed on a monitor one at a time with the active camera selected by voluntary blinking. EOG signals were converted to pixel coordinates using an affine transform. The pixel coordinates were then transformed into robot coordinates using the correlation between camera position, robot position and monitor display. Robot coordinates of a target became end-effector position and joint angles were derived through inverse kinematics. The result shows that this system could reach target with approximately 90% accuracy. This system successfully improved the flexibility of controlling a robot in three dimensions.

show abstract

“…Başkent Üniversitesinde yapılan göz hareketleriyle ilgili yazı yazma çalışmasında dikey ve yatay EOG sinyalleri kodlanarak 42 karakter oluşturulmuştur [11]. EOG sinyallerinin sınıflandırılması için yapılan çalışmada destek vektör makinesi ve yapay sinir ağları yöntemleri karşılaştırılmıştır [12][13].…”

Section: Introductionunclassified

EOG based intelligent direction detect system with pre-filtering algorithm

Erkaymaz

Özer

Kaya

et al. 2015

2015 23nd Signal Processing and Communications Applications Conference (SIU)

View full text Add to dashboard Cite

Özetçe-Günümüzde el, ayak veya kollarını kullanamayan hastalar için sorunlu organlar yerine diğer organlardan faydalanarak yapay hareketler elde edilebilmektedir. Bu bağlamda göz hareketlerinin dış ortamlara mesaj verebilmek için kullanımı popüler bir bilimsel araştırma konusu olmuştur. Göz hareketlerine bağlı olarak yapılan çalışmalarda elektriksel kökenli biyolojik işaret olan elektrookulogram (EOG) sinyalinden yararlanılmaktadır. Bu çalışmada gözün etrafına yerleştirilen elektrotlar vasıtasıyla alınan yatay ve dikey EOG sinyalleri yapay zekâ tekniklerinden Yapay Sinir Ağları (YSA) yöntemi ile modellenmiştir. Sistem; sağa, sola, yukarı, aşağı olmak üzere dört ana yönü tespit edebildiği gibi göz kırpma hareketlerini de tespit edebilmektedir. Alınan sinyaller ilk olarak bilgisayarda ön eşik filtreleme ve yükselteç işlemlerinden geçirilip, YSA ile sınıflandırılmıştır. Hesaplanan özelliklere göre gelen sinyaller ile önerilen modelinin performansı istatistiksel doğruluk ve karışıklık matrisi analizleri yapılarak ortaya konulmuştur. Analiz sonuçlarında görülmüştür ki, tasarlanan model göz hareketlerini başarılı bir şekilde tespit edebilmektedir. Abstract-Nowadays, artificial movements have been obtained by utilizing other organs for paralyzed patients.Especially the usage of eye movements for giving message to outside world became popular as a scientific subject. In studies according to eye movements, the Electrooculogram (EOG) signal is used. In this study, the vertical and horizontal EOG signals taken from electrodes, placed around the eyes, have been modelled by using Artificial Neural Networks (ANN) which is one of artificial intelligent technique. The system can sense four main directions (Right, Left, Up and Down) at the same time it can also detect blinking movements. Firstly, the signals have been pre-filtered, amplified and classified by ANN. The performance of recommended model has been demonstrated by analyzing statistical accuracy and confusion matrix according to the features of obtained signal. It has been seen that eye movements can be successfully determined by designed model.

show abstract

Classifying Electrooculogram to Detect Directional Eye Movements

Cited by 40 publications

References 13 publications

Electrooculogram based blink detection to limit the risk of eye dystonia

Electrooculogram based blink detection to limit the risk of eye dystonia

Controlling 3-D Movement of Robot Manipulator using Electrooculography

EOG based intelligent direction detect system with pre-filtering algorithm

Contact Info

Product

Resources

About